[time-nuts] Advantages & Disadvantages of the TPLL Method

Magnus Danielson magnus at rubidium.dyndns.org
Tue Jun 15 19:16:08 UTC 2010


Hi Warren,

On 06/15/2010 08:49 PM, WarrenS wrote:
> Magnus posted a bunch of good stuff,
>
>> Neither. It's a characteristic, it needs to be analyzed. If the DUT is
>> very sensitive, then additional care may be taken or maybe it just isn't
>> a very good solution.
>
> We have little disagreements for the most part. Except maybe for if it
> should be on 'the list'.
> The characteristics are what makes the advantages and disadvantages, so
> they belong on the list.

Well, it depends on the oscillators involved... so it is just not 
given... but assuming a fairly typical oscillator for our world, then it 
would end up on the advantage side.

> The high natural rejection that the TPLL has to injection locking makes
> it much less sensitive than any other system, so it lets one keep it
> simple. TO me that is an advantage.

Hmm, yes... for the typical oscillator yes...

> I did test for the effect by connecting the REF osc to the DUT osc with
> a variable attenuator, and could not make the TPLL system start to fail
> until I the coupling got so low that it started to short out the two
> signals, and still it was working, just got a bit noisier
> I called the results of that test good enough and moved on to other ways
> of testing.
> All of which it also passed with flying colors

Noiser as you added gain...

> Another subtle side effect of that advantages is that the cables do not
> to be shielded so well.
> example:
> If I bring a third osc's that is connect to a RG58 BNC shielded cable
> and is offset in freq by 1 to 10 Hz
> The effect of coupling (and or Injection locking) between to BNC
> shielded cables can be seem at the Tester output.
>
> What I have found is if I want to get the full accuracy of the TPLL, I
> can not have an offset osc cable on the same table, unless it is fully
> enclosed in a RF proof box.

That would help, definitly.

Cheers,
Magnus




More information about the Time-nuts_lists.febo.com mailing list